Fetch MCP Rust

What it does

Fetch MCP Rust is a high-performance web content acquisition server that allows AI agents to interact with the live web. It transforms messy HTML into clean, LLM-friendly Markdown and extracts structured data from various web sources including RSS feeds, sitemaps, and specific API integrations like Reddit and Wikipedia.

Tools

• fetch: Converts URL content to clean Markdown using a Readability algorithm.
• fetch_metadata: Extracts Open Graph, Schema.org, and HTML meta tags.
• fetch_feed: Parses RSS/Atom feeds and extracts entries.
• fetch_with_selector: Extracts specific HTML elements using CSS selectors.
• extract_table: Converts HTML tables into structured JSON.
• fetch_sitemap: Parses sitemap.xml files to discover page URLs.
• fetch_links: Extracts internal and external links from a page.
• fetch_batch: Performs parallel URL fetching with built-in rate limiting.
• search_in_page: Searches for specific text within a page with context extraction.
• reddit: Searches Reddit posts and extracts comments.
• wiki: Fetches summaries or full articles from Wikipedia.
• fetch_pdf_text: Extracts text from PDF documents.
• fetch_image_info: Retrieves image dimensions and format.

Installation

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "fetch": {
      "command": "/path/to/fetch-mcp-rs",
      "args": []
    }
  }
}

Supported hosts

• Claude Desktop